Sirolimus and Thrombotic Microangiopathy after Allogeneic Stem Cell Transplantation.

Sirolimus (SRL) is a novel immunosuppressive agent that has been demonstrated to reduce GVHD and minimize morbidity after allogeneic stem cell transplantation (alloSCT). We have described a syndrome of thrombotic microangiopathy (TMA), characterized by microangiopathic hemolysis, thrombocytopenia, and renal dysfunction, whose incidence is increased when SRL is used in association with calcineurin inhibitors (CIs) but not with SRL monotherapy. To determine if SRL use potentiates the effects of CIs on TMA incidence and risk factors, we performed a retrospective cohort analysis of subjects who underwent alloSCT between 1997 and 2003.

Methods: Subjects who received a SRL-containing GVHD prophylaxis after a myeloablative preparative regimen were compared with a cohort who received a non-SRL regimen. All subjects received CIs. Diagnosis of TMA required the simultaneous occurrence of: (1) creatinine elevation >2 mg/dL or >50% above baseline, (2) schistocytosis, (3) elevated LDH, and (4) no laboratory evidence of disseminated intravascular coagulopathy.

Results: 111 patients who received SRL were compared with 216 patients who received no SRL during the first 100 days after alloSCT. The two groups of patients were balanced for demographic parameters; however, more patients in the SRL group received peripheral blood stem cells (50.5 vs. 18.1%, p<0.01) and had unrelated donors (58.6 vs. 42.6%, p<0.01). The incidence of TMA in the SRL group was 10.8% in comparison with an incidence of 4.2% in the non-SRL group (OR 2.57, p=0.03). Patients who received SRL developed TMA earlier than those who did not receive SRL (median 25 vs. 58 days, p=0.04). At the time of TMA diagnosis, median blood levels of immunosuppressive medications were in their respective therapeutic ranges: SRL (study group) 6.1 ng/ml, tacrolimus (study group) 9.9 ng/ml, tacrolimus (control group) 9.1 ng/ml; cyclosporine (control group) 418 ng/ml. In a multivariable logistic regression model, only the use of SRL (Adjusted Exact OR 3.49, p =0.02) and grade II-IV acute GVHD (Adjusted Exact OR 6.60, p = 0.0002) predicted the occurrence of TMA. Treatment of TMA consisted of discontinuation or dose adjustment of CIs. SRL was discontinued or dosed according to serum level. Two subjects in each group required temporary hemodialysis, and 3 subjects (1 SRL, 2 non-SRL) underwent plasmapheresis. 78% of surviving SRL-treated subjects regained normal renal function. No subject had a TMA recurrence if SRL was reintroduced. Overall survival after TMA diagnosis was better for SRL patients than non-SRL patients (58.3 vs. 11.1%, log rank p=0.02).

Conclusion: SRL use is associated with an increased risk of TMA after alloSCT and may act by potentiating the effects of CIs. TMA associated with SRL appears reversible and does not affect overall survival after alloSCT. A careful monitoring strategy for TMA should be employed as part of a SRL-containing GVHD prophylaxis regimen.